KR101019547B1 - Iterative multipath interface cancellation system and method - Google Patents

Abstract

A system and method for Interference Cancellation (IC). One aspect relates to iterative interference cancellation with iterative finger delay adaptation. The interference cancellation method comprises receiving multi-paths of a signal; and performing iterative interference cancellation to remove multi-path interference, wherein the performing iterative IC comprises estimating a Signal-to-Interference-plus-Noise Ratio (SINR) at each of a plurality of pre-determined rake receiver finger delays, and performing successive Channel Estimation (CE) and IC on rake receiver fingers according to their estimated SINRs, and wherein the CE of a next finger does not start until interference of a previous finger is removed from a sample buffer. The method may further comprise improving estimated rake receiver finger delay, and each iteration decreases the amount of interference observed by each finger.

Description

Iterative multipath interference cancellation system and method {ITERATIVE MULTIPATH INTERFACE CANCELLATION SYSTEM AND METHOD}

Priority Claim Under 35 USC §119 This application claims priority to preliminary application 60 / 730,224, filed October 24, 2005, entitled “Repeat Interference Cancellation,” which is incorporated herein by reference.

FIELD OF THE INVENTION The present invention generally relates to communication systems, and in particular, to repetitive interference cancellation in wireless communication systems.

The communication system may provide communication between a base station and access terminals (AT). The forward link or downlink is associated with the transmission from the base station to the access terminal. The reverse link or uplink is associated with the transmission from the access terminal to the base station. Each access terminal may communicate with one or more base stations on the forward and reverse links at any instant, depending on whether the access terminal is active and whether the access terminal is in soft handoff.

Interference Cancellation (IC) systems and methods are provided. One form relates to repetitive interference cancellation by repetitive finger delay adaptation. An interference cancellation method includes the steps of receiving a signal of multiple paths; And performing iterative interference cancellation to remove multipath interference, wherein the iterative IC performing step estimates the signal-to-interference + noise ratio (SINR) at each of a plurality of predetermined rake receiver fingers having a predetermined delay. And performing continuous channel estimation (CE) and IC according to the estimated SINRs of the fingers on the rake receiver fingers, wherein the next finger until the interference of the previous finger is removed from the sample buffer. CE does not start. The method also includes improving the estimated rake receiver finger delay using the previous structure at each finger.

The features, characteristics and advantages of the present application may become more apparent from the following detailed description taken in conjunction with the drawings. The same reference numbers and symbols may identify the same or similar objects.
1 shows a wireless communication system having a base station and access terminals.

2 illustrates an example of a transmitter structure and / or process that may be implemented in the access terminal of FIG. 1.

3 illustrates an example of a receiver process and / or structure that may be implemented in the base station of FIG. 1.

4 illustrates another embodiment of a base station receiver process or structure.

5 shows a general example of power distribution of three users in the system of FIG. 1.

6 shows an example of even time offset distribution for frame asynchronous traffic interference cancellation for users with the same transmit power.

7 illustrates an interlacing structure used for a reverse link data packet and a forward link auto repeat request channel.

8 shows a memory spanning a complete 16-slot packet.

9A illustrates a method of traffic interference cancellation for an example of sequential interference cancellation (SIC) without decoding delay.

9B illustrates an apparatus for performing the method of FIG. 9A.

10 shows a receiver sample buffer after arrival of successive subpackets of an interlace with interference cancellation (IC) of decoded subpackets.

11 shows an overhead channel structure.

12A illustrates a method for first performing a pilot IC (PIC) and then performing an overhead IC (OIC) and a traffic IC (TIC) together.

12B illustrates an apparatus for performing the method of FIG. 12A.

13A illustrates a variation of the method of FIG. 12A.

FIG. 13B illustrates an apparatus for performing the method of FIG. 13A.

14A shows a method for performing joint PIC, OIC and TIC.

FIG. 14B illustrates an apparatus for performing the method of FIG. 14A.

15A illustrates a variation of the method of FIG. 14A.

FIG. 15B illustrates an apparatus for performing the method of FIG. 15A.

16 shows a model of a transmission system.

17 illustrates an example response of combined transmit and receive filtering.

18A and 18B show examples of channel estimation (real and imaginary components) based on estimated multipath channels at each of the three rake fingers.

19A and 19B show an example of an improved channel estimate based on rake fingers and despread by data chips.

20A illustrates a method of despreading at Rake finger delay with regenerated data chips.

20B shows an apparatus for performing the method of FIG. 20A.

21A and 21B show an example of estimating a composite channel using evenly spaced samples at chipx2 resolution.

22A illustrates a method of estimating a composite channel at a constant resolution using regenerated data chips.

22B illustrates an apparatus for performing the method of FIG. 22A.

23 illustrates closed loop power control and gain control with a constant overhead subchannel gain.

24 is a variation of power control and gain control due to the constant overhead subchannel gain of FIG.

25 shows an example of power control with a constant overhead subchannel gain.

FIG. 26 is similar to FIG. 24 except for overhead gain control.

FIG. 27 shows a modification of FIG. 26 by DRC dedicated overhead gain control.

28 shows a graph of the actual and reconstructed channel impulse response (CIR).

29A illustrates a method for iterative finger delay adaptation.

FIG. 29B illustrates an apparatus for performing the method of FIG. 29A.

30 shows a graph of actual, reconstructed and improved CIRs.

Any embodiment described herein is not necessarily preferred or advantageous over other embodiments. While various forms of the disclosure are set forth in the drawings, the drawings are not necessarily drawn to scale or to be all-inclusive.

1 illustrates a wireless communication system 100 that includes a system, a controller 102, base stations 104a and 104b, and a number of access terminals 106a-106h. System 100 may have any number of system controllers 102, base stations 104, and access terminals 106. Various aspects and embodiments of the present disclosure described below can be implemented in the system 100.

The access terminal 106 may be mobile or fixed and may be scattered throughout the communication system 100 of FIG. 1. The access terminal 106 can be connected to or implemented with a computing device, such as a laptop personal computer. Alternatively, the access terminal may be a standalone data device such as a personal digital assistant (PDA). Access terminal 106 may be associated with various types of devices, such as wired telephones, wireless telephones, cellular phones, laptop computers, wireless communications personal computer (PC) cards, PDAs, external or internal modems, and the like. An access terminal may be any device that provides data connectivity to a user by communicating over a wireless channel or over a wired channel, for example using an optical fiber or coaxial cable. The access terminal may have various names such as mobile station, access unit, subscriber unit, mobile device, mobile terminal, mobile unit, mobile phone, mobile, remote station, remote terminal, remote unit, user device, user equipment, handheld device, and so on. have.

System 100 provides communication for a number of cells, each cell being serviced by one or more base stations 104. Base station 104 may be referred to as a base station transceiver system (BTS), an access point, part of an access network (AN), a modem pool transceiver (MPT), or a Node B. An access network is associated with network equipment that provides data connectivity between a packet switched data network (eg, the Internet) and an access terminal 106.

The forward link (FL) or downlink is related to the transmission from the base station 104 to the access terminal 106. The reverse link (RL) or uplink is related to the transmission from the access terminal 106 to the base station 104.

Base station 104 may transmit data to access terminal 106 using a data rate selected from a set of different data rates. The access terminal 106 can measure the signal-to-interference + noise ratio (SINR) of the pilot signal transmitted by the base station 104 and determine the desired data rate for the base station 104 to transmit data to the access terminal 106. . The access terminal 106 may send a data request channel or data rate control (DRC) message to the base station 104 to inform the base station 104 of the desired data rate.

The system controller 102 (also referred to as a base station controller (BSC)) can provide coordination and control for the base station 104 and also control the routing of calls to the access terminal 106 by the base station 104. Can be. The system controller 102 may also be connected to a public switched telephone network (PSTN) via a mobile switching center (MSC) and to a packet data network through a packet data serving node (PDSN).

Communication system 100 is a Code Division Multiple Access (CDMA), IS-95, high data rate is referred to as (HDR), "CDMA2000 ^® High Rate Packet Data Air Interface Specification," the high-speed packet data described in detail (HRPD), TIA / EIA / IS-856, CDMA 1x Optimized Deployment Data (EV-DO), 1xEV-DV, Wideband CDMA (WCDMA), Universal Mobile Telecommunications System (UMTS), Time Division Synchronous CDMA (TD-SCDMA), Orthogonal Frequency Division Multiplexing One or more communication techniques, such as (OFDM), may be used. The following examples provide details for clarity of understanding. The ideas presented herein may be applied to other systems, and the examples are not intended to limit the present application.

2 illustrates an example of a transmitter structure and / or process that may be implemented in the access terminal 106 of FIG. The functions and components shown in FIG. 2 may be implemented in software, hardware, or a combination of software and hardware. In addition to or instead of the functions shown in FIG. 2, other functions may be added to FIG. 2.

Data source 200 provides data to encoder 202, which encodes data bits using one or more schemes to provide coded data chips. Each coding scheme may include one or more coding types, including cyclic redundancy check (CRC), convolutional coding, turbo coding, block coding, other types of coding, or no coding at all. Other coding schemes may include automatic repeat request (ARQ), hybrid ARQ (H-ARQ), and progressive redundancy iteration techniques. Different types of data may be coded with different coding schemes. Interleaver 204 interleaves the coded data bits to deal with fading.

Modulator 206 modulates the coded interleaved data to produce modulated data. Examples of modulation techniques include binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK). The modulator 206 may also repeat the sequence of modulated data or the symbol puncture unit may puncture the bits of the symbol. The modulator 206 may also spread the modulated data by Walsh cover (ie, Walsh code) to form data chips. The modulator 206 may also time division multiplex the data chips with pilot chips and MAC chips to form a chip stream. The modulator 206 can also spread the chip stream into one or more PN codes (eg, short code, long code) using a pseudo random noise (PN) spreader.

The baseband-to-radio frequency conversion unit 208 may convert the baseband signal into a radio frequency (RF) signal for transmission by the antenna 210 to one or more base stations 104 via a wireless communication link.

3 illustrates an example of a receiver process and / or structure that may be implemented in the base station 104 of FIG. The functions and components shown in FIG. 3 may be implemented in software, hardware, or a combination of software and hardware. In addition to or instead of the functions shown in FIG. 3, other functions may be added to FIG. 3.

One or more antennas 300 receive reverse link modulated signals from one or more access terminals 106. Multiple antennas can provide spatial diversity against harmful path effects, such as fading. Each received signal is provided to a separate receiver or RF-baseband conversion unit 302 to adjust (eg, filter, amplify, downconvert) the received signal and digitize it to generate data samples for that received signal.

The demodulator 304 can demodulate the received signals and provide recovered symbols. For CDMA2000 ^®, demodulation (1) screen despread samples of channelization to isolate the received data and pilot on each code channel, or channels, (2) a pilot restored the channelized data coherent demodulation Attempt to restore data transmission by providing demodulated data. Demodulator 304 despreads and processes multiple signal instances, a receive sample buffer 312 (also called joint front-end RAM (FERAM) or sample RAM) that stores samples of received signals for all user / access terminals. A rake receiver 314 and a demodulated symbol buffer 316 (also called a back end RAM (BERAM) or demodulated symbol RAM). There may be multiple demodulation symbol buffers 316 corresponding to multiple users / access terminals.

Deinterleaver 306 deinterleaves data from demodulator 304.

The decoder 308 may decode the demodulated data to recover the decoded data bits sent by the access terminal 106. The decoded data may be provided to the data sink 310.

4 illustrates another embodiment of a base station receiver process or structure. In FIG. 4, the data bits of the successively decoded user are input to the interference reconstruction unit 400, which interpolates the encoder 402, the interleaver 404, the modulator 406 and the filter 408. Include. Encoder 402, interleaver 404, and modulator 406 may be similar to encoder 202, interleaver 204, and modulator 206 of FIG. 2. The filter 408 forms samples of the user decoded with FERAM resolution, for example with a change from chip rate to 2x chip rate. The decoder user's contribution to FERAM is removed or deleted at FERAM 312.

Although interference cancellation at base station 104 is described below, the concepts herein can be applied to access terminal 106 or any other component of a communication system. Traffic Interference Rejection

Since signals transmitted by different users are not orthogonal at the base station or BTS 104, the capacity of the CDMA reverse link may be limited by interference between users. Therefore, techniques that reduce interference between users will improve system performance of the CDMA reverse link. The technology for the efficient implementation of interference cancellation for advanced CDMA systems such as CDMA2000 ^® 1xEV-DO RevA is described herein.

Each DO RevA user transmits traffic, pilot and overhead signals, all of which can cause interference to other users. As FIG. 4 shows, signals at BTS 104 may be reconstructed and subtracted at FERAM 312. The transmitted pilot signal is known to the BTS 104 and can be reconstructed based on the information about the channel. However, overhead signals (such as reverse rate indicator (RRI), data request channel or data rate control (DRC), data source channel (DSC), acknowledgment (ACK)) are first demodulated and detected and transmitted in BTS 104. The demodulated data signals are demodulated, deinterleaved and decoded to determine the overhead and traffic chips transmitted. Based on the determination of the chips transmitted for a given signal, the reconstruction unit 400 may reconstruct the contribution to the FERAM 312 based on the channel information.

The bits of the data packet from data source 200 may be repeated and a number of corresponding “subpackets” for transmission to BTS 104 by encoder 202, interleaver 204 and / or modulator 206. Can be treated as If the BTS 104 receives a high signal-to-noise ratio signal, the first subpacket may contain enough information for the BTS 104 to decode and derive the original data packet. For example, data packets from data source 200 may be repeated and processed into four subpackets. The user terminal 106 sends the first subpacket to the BTS 104. The BTS 104 may have a relatively low probability of correctly decoding and deriving the original data packet from the received first subpacket. However, when the BTS 104 receives the second, third and fourth subpackets and combines the information derived from each received subpacket, the probability of decoding and deriving the original data packet is high. As soon as the BTS 104 correctly decodes the original packet (e.g., using a CRC or other error detection technique), the BTS 104 sends an acknowledgment signal to the user terminal 106 to stop transmission of the subpackets. . The user terminal 106 can then send the first subpacket of the new packet.

The reverse link of DO-RevA uses H-ARQ (FIG. 7), where each 16-slot packet is broken down into four subpackets and transmitted in an interlace structure with eight slots between subpackets of the same interlace. Moreover, different user / access terminals 106 may initiate their transmissions at different slot boundaries, such that four-slot subpackets of different users arrive at the base station or BTS asynchronously. The effect of efficient design and asynchronousness of interference cancellation receivers on H-ARQ and CDMA is described later.

The gain from interference cancellation depends on the order in which signals are removed from the FERAM 312. Techniques related to decoding (and subtracting if passing a CRC) of users based on traffic-to-pilot (T2P) ratio, effective SINR or decoding probability are disclosed herein. Various approaches are disclosed herein to retry demodulation and decoding of users after other users are removed from the FERAM 312. Interference cancellation from the BTS FERAM 312 can be efficiently implemented in consideration of an asynchronous CDMA system such as EV-DO RevA, and users use a hybrid ARQ to transmit pilot signals, control signals and traffic signals. The present disclosure may be applied to EV-DV Rel D, W- CDMA EUL , and CDMA2000 ^®.

Traffic interference cancellation (TIC) may be defined as subtraction interference cancellation, which removes the contribution of user data to FERAM 312 after the user has been correctly decoded (FIG. 4). Some of the practical problems associated with TIC on actual CDMA systems such as ^{CDMA2000 ®, EV-DO, EV} -DV , and WCDMA are described here. Many of these problems are due to the fact that real systems have user asynchronous and H-ARQ. For example, CDMA2000 ^® is uniformly diffused in time user data frames to prevent excess delay in the backhaul network. RevA of EV-DO, Rel D of EV-DV, and EUL of WCDMA also use H-ARQ to derive possible data lengths greater than one.

Multi-user detection is the main category of algorithms in which TICs are made and is associated with any algorithm that seeks to improve performance by allowing detection of two different users to interact. The TIC method may include a mixture of continuous interference cancellation or sequential interference cancellation (SIC) and parallel interference cancellation. The STC involves any algorithm that sequentially decodes users and uses the data of previously decoded users to improve performance. PIC is generally concerned with decoding users simultaneously and excluding all decoded users at the same time.

The TIC may be different from the PIC. One difference between TIC and PIC is that the transmitted pilot signal is completely known by the receiver. Thus, the PIC can subtract the pilot contribution to the received signal using only the channel estimates. The second major difference is that the transmitter and receiver interact closely on the traffic channel via the H-ARQ mechanism. The receiver does not know the transmitted data sequence until the user has been successfully decoded.

Similarly, it is desirable to delete overhead channels from FERAM in a technique called overhead interference cancellation (OIC). Overhead channels cannot be deleted until either the base station or the BTS 104 knows the transmitted overhead data, which is determined by decoding and then correcting the overhead message.

Continuous interference cancellation defines a class of methods. The chain law of mutual information shows that under ideal conditions, continuous interference cancellation can achieve the capacity of multiple access channels. The main conditions for this are that all users are frame synchronized and can be estimated with errors that each user's channel can ignore.

5 shows a general example of power distribution of three users (User 1, User 2, User 3), where users transmit frames simultaneously (frames from all users are received at the same time) and each user has the same data. It is transmitting at the rate. Each user is instructed to use a particular transmit power, for example User 3 transmits at about the same power as noise; User 2 transmits at about the same power as User 3's power plus noise; User 1 transmits at about the same power as User 2 + User 3 + Noise.

The receiver processes signals from users in order of decreasing by transmit power. Starting at k = 1 (user 1 with the highest power), the receiver attempts to decode user 1. If decoding is successful, User 1's contribution to the received signal is formed and subtracted based on its channel estimate. This may be referred to as frame synchronization sequential interference cancellation. The receiver continues until decoding is attempted for all users. Each user has the same SINR after successive interference cancellation of previously decoded users.

Unfortunately, this approach can be very sensitive to decoding errors. If one large power user such as user 1 is not decoded correctly, the SINR of all subsequent users may be severely degraded. This may prevent the decoding of all users after that point. Another drawback of this approach is that it requires users to have a specific relative power at the receiver, which is difficult to guarantee in a fading channel. Frame asynchronous and interference cancellation, e.g. CDMA2000 ^®

Assume that user frame offsets are intentionally staggered. This frame asynchronous operation has many benefits for the system as a whole. For example, processing power and network bandwidth at the receiver will have a more even usage profile in time. In contrast, frame synchronization between users requires a burst of processing power and network resources at the end of each frame boundary because all users complete the packet at the same time. By frame asynchronous, the BTS 104 can decode the user with the earliest arrival time first rather than the user with the highest power.

6 shows an example of even time offset distribution of frame asynchronous TICs for users with the same transmit power. 6 shows a snapshot just before frame 1 of user 1 is to be decoded. Since frame 0 for all users has already been decoded and deleted, the contribution to the interference is represented by cross parallel lines (users 2 and 3). In general, this approach reduces interference by a factor of two. Half of the interference was eliminated by the TIC before decoding Frame 1 of User 1.

In another embodiment, the users of FIG. 6 may be associated with a group of users, eg, user group 1, user group 2, user group 3.

The benefit of asynchronous and interference cancellation is the relative symmetry between users if they want similar data rates in terms of power levels and error statistics. In general, by sequential interference cancellation with the same user data rate, the last user is received at very low power and also relies heavily on the successful decoding of all previous users. Asynchronous, hybrid ARQ and interlacing , for example EV-DO RevA

7 shows an interlacing structure (eg, of 1xEV-DO RevA) used for RL data packet and FL ARQ channel. Each interlace (interlace 1, interlace 2, interlace 3) contains a set of segments staggered in time. In this example, each segment is four time slots long. During each segment, the user terminal may send a sub packet to the base station. There are three interlaces, and each segment is four time slots long. Thus, there are eight time slots between the end of a subpacket of a given interlace and the start of the next subpacket of the same interlace. This provides enough time for the receiver to decode the subpacket to relay an ACK or negative acknowledgment (NAK) to the transmitter.

Hybrid ARQ takes advantage of the time varying nature of fading channels. If the channel condition for the first one, two or three subpackets is good, the data frame can be decoded using only these packets and the receiver sends an ACK to the transmitter. The ACK does not send the remaining subpacket (s) but directs the transmitter to start a new packet if desired. Receiver structure for interference cancellation

By the TIC, the data of the decoded users are reconstructed and subtracted (FIG. 4) so that the BTS 104 can remove the interference caused by the data of the decoded users to other users. The TIC receiver may have two circular memories: FERAM 312 and BERAM 316.

FERAM 312 stores the received samples (eg, at 2x chip rate), which is common to all users. Non-TIC receivers use only about 1-2 slots of FERAM (to adjust delay in the demodulation process) because no deduction of traffic or overhead interference occurs. In a TIC receiver for a system with H-ARQ, the FERAM can span many slots, for example 40 slots, and is updated by the TIC through the exclusion of decoded users. In another configuration, the FERAM 312 may have a length shorter than the entire packet, such as the length of the period from the start of the subpacket of the packet to the end of the next subpacket of the packet.

BERAM 316 stores demodulated symbols of the received bits as generated by the rake receiver 314 of the demodulator. Each user may have a different BERAM because the demodulated symbols are obtained by despreading by a user specific PN sequence and combining rake fingers. Both TIC and non-TIC receivers may use BERAM 316. In the TIC, BERAM 316 is used to store demodulated symbols of previous subpackets that are no longer stored in FERAM 312 when the FERAM 312 does not span all subpackets. BERAM 316 may be updated each time a decoding attempt is made or whenever a slot exists from FERAM 312. How to choose FERAM length

The size of the BERAM 316 and FERAM 312 may be selected according to various trade-offs between the required processing power of the system, the transmission bandwidth from memory to the processor, delay and performance. In general, the use of shorter FERAM 312 will limit the gain of the TIC since the oldest subpacket will not be updated. On the other hand, the shorter FERAM 312 yields a reduced number of demodulations, subtractions, and lower transmission bandwidths.

By RevA interlacing, 16-slot packets (four subpackets each transmitted in four slots) span 40 slots. Thus a 40-slot FERAM can be used to ensure removal of the user from all affected slots.

8 shows a 40-slot FERAM 312 spanning a complete 16-slot packet for EV-DO RevA. Each time a new subpacket is received, decoding is attempted on that packet using all available subpackets stored in FERAM 312. If decoding is successful, the contribution of that packet from FERAM 312 is deleted by reconstructing and subtracting the contribution of all component subpackets 1, 2, 3 or 4. The DO-RevA FERAM length of 4, 16, 28 or 40 slots can span 1, 2, 3 or 4 subpackets, respectively. The length of the FERAM implemented at the receiver may depend on complexity considerations, the need to support various user arrival times, and the ability to redo demodulation and decoding of users at previous frame offsets.

9A illustrates a general TIC method for an example of sequential interference cancellation without a decoding delay. Other improvements are described below. The process begins at start block 900 and proceeds to delay selection block 902. In the SIC, the delay selection block 902 may be omitted. In block 903, the BTS 104 selects one of the users (or user group) at which the sub packet ends in the current slot.

In block 904, demodulator 304 demodulates samples of the user's subpackets for some or all time segments stored in FERAM 312 according to the user's spreading and scrambling sequence as well as its arrangement size. At block 906, the decoder 308 attempts to decode the user packet using previous demodulated symbols and demodulated FERAM samples stored in the BERAM 316.

At block 910, the decoder 308 or other unit may determine whether the packet of the user (s) has successfully decoded, that is, passes an error check such as using a CRC.

If the user packet fails to decode, the NAK is sent back to the access terminal 106 at block 918. If the user packet is decoded correctly, an ACK is sent to the access terminal 106 at block 908 and interference cancellation is performed at blocks 912-914. Block 912 regenerates the user signal according to the decoded signal, channel impulse response, and transmit / receive filter. Block 914 subtracts the user's contribution from the FERAM 312 to eliminate interference for users that have not yet been decoded.

On both failure and success of decoding, the receiver moves to the next user to be decoded at block 916. If decoding attempts have been made for all users, a new slot is inserted in FERAM 312 and the entire process is repeated for the next slot. Samples are recorded in FERAM 312 in real time, ie 2x chip rate samples may be recorded every half chip.

9B shows an apparatus comprising means 930-948 for performing the method of FIG. 9A. The means 930-948 of FIG. 9B may be implemented in hardware, software, or a combination of hardware and software. How to choose the decoding order

Block 903 indicates that the TIC may be applied sequentially to each user or in parallel to a group of users. As the group grows, the implementation complexity may decrease but the gain of the TIC may decrease as long as the TIC is not repeated as described below.

The criteria by which users are grouped and / or sorted may vary depending on the channel variation rate, traffic type and available processing power. A good decoding order may initially include decoding of users whose deletion is most useful and most easy to decode. Criteria for achieving the greatest benefit from TIC may include:

A. Payload Size and T2P: The BTS 104 can group or sort users according to payload size and decode them in order starting with the lowest T2P at the highest transmit power, i.e. users with the highest T2P. can do. Decoding and deletion of high T2P users from FERAM 312 has the greatest benefit since they cause the most interference to other users.

B. SINR: The BTS 104 may decode users with higher SINRs before users with lower SINRs because users with higher SINRs have higher decoding probabilities. In addition, users with similar SINRs can be grouped together. For fading channels, the SINR varies over time throughout the packet so that an equivalent SINR can be calculated to determine the proper order.

C. Time: BTS 104 may decode “older” packets (ie, packets where more subpackets have been received at BTS 104) before “newer” packets. This selection reflects the assumption that packets are easier to decode into each incremented subpacket for a given T2P ratio and ARQ termination purpose. Retry method of decoding

The interference contribution is subtracted from the FERAM 312 when the user is correctly decoded, increasing the potential to correctly decode all users sharing some slots. Since the interference perceived by previously failed users has dropped considerably, it is advantageous to repeat the decoding attempts of these users. Delay selection block 902 selects the slot (current or past) used as a reference for decoding and IC. The user selection block 903 will select the users whose subpackets end in the slot of the selected delay. The choice of delay can be based on the following options:

A. If decoding has been attempted for all users, then the current decoding indicates a move selection to the next (future) slot, and the next slot may be used for FERAM 312. In this case, decoding of each user is attempted once for each slot to be processed, which corresponds to continuous interference cancellation.

B. Iterative decoding attempts to decode the users more than once per slot processed. The second and next decoding iterations will benefit from the removed interference of users decoded in the previous iterations. Iterative decoding yields gains when multiple users are decoded in parallel without intervening ICs. With purely iterative decoding on the current slot, delay selection block 902 simply selects the same slot (ie, delay) multiple times.

C. Backward Decoding: The receiver demodulates the subpackets and attempts to decode the packet based on the demodulation of all available subpackets of the FERAM corresponding to that packet. After attempting to decode packets with subpackets ending in the current time slot (ie, users at the current frame offset), the receiver attempts to decode packets that failed to decode in the previous slot (ie, users at the previous frame offset). can do. Due to partial overlap between asynchronous users, the eliminated interference of subpackets ending in the current slot will improve the decoding chance of previous subpackets. The process can be repeated by returning to more slots. The maximum delay of the forward link ACK / NAK transmission may limit backward decoding.

D. Forward decoding: After attempting to decode all packets having subpackets ending in the current slot, the receiver may attempt to decode the most recent users before the entire subpacket is written to FERAM. For example, the receiver could attempt to decode users after three of the four slots of the most recent subpacket have been received. How to update BERAM

In a non-TIC BTS receiver, packets are decoded based only on the demodulated symbols stored in BERAM, and FERAM is only used to demodulate users from the most recent time segments. By TIC, the FERAM 312 is still accessed whenever the receiver attempts to demodulate a new user. However, by the TIC, the FERAM 312 is updated after the user is correctly decoded based on the reconstruction and subtraction of that user's contribution. Due to complexity considerations, it may be desirable to choose a FERAM buffer length shorter than the packet's distance (eg, 40 slots are needed to span 16-slot packets in EV-DO RevA). When new slots are written to FERAM 312, they overwrite the oldest samples in the circular buffer. Thus, when new slots are received, the oldest slots will be overwritten and decoder 308 will use BERAM 316 for these old slots. Even if a given subpacket is not located in FERAM 312, BERAM 316 stores the most recent demodulated (determined from FERAM 312) symbols of the demodulator for that subpacket as an intermediate step in the interleaving and decoding process. It should be noted that it can be used to

A. User Based Updates: The BERAM 316 for a user is updated only in relation to the decoding attempted for that user. In this case, the update of older FERAM slots may not gain the benefit of BERAM 316 for that user if the given user is not decoded in a timely manner (ie, updated FERAM slots are not FERAM before the user's decoding is attempted). Exit 312).

B. Slot Based Update: To fully utilize the benefits of TIC, BERAM 316 for all affected users can be updated each time a slot exits FERAM 312. In this case, the contents of BERAM 316 include all interference subtractions made to FERAM 312. How to remove interference from arrived subpackets due to missed ACK deadlines

In general, the extra processing used by the TIC introduces a delay in the decoding process, which is particularly appropriate when an iterative or backward approach is used. This delay may exceed the maximum delay at which an ACK can be sent to the transmitter to stop transmission of subpackets related to the same packet. In this case, the receiver can still use successful decoding by using the decoded data to subtract not only previous subpackets but also subpackets that will be received in the near future due to the missing ACK.

By the TIC, the data of the decoded users can be reconstructed and subtracted to remove interference caused by the base station 104 in sub-packets of other users. By H-ARQ, whenever a new subpacket is received, decoding of the original packet is attempted. If decoding is successful, the contribution of that packet from the received samples can be removed by reconstructing and subtracting the component subpackets for H-ARQ by the TIC. Depending on the complexity considerations, interference can be eliminated from one, two, three or four subpackets by storing a longer history of samples. In general, the IC may be applied sequentially to each user or to a group of users.

10 shows a receiver sample buffer 312 at three time instances: slot time n, n + 12 slots and n + 24 slots. For illustration, FIG. 10 shows a single interlace with subpackets from three users on the same frame offset to emphasize interference cancellation by H-ARQ. Receiver sample buffer 312 of FIG. 10 spans all four subpackets (which can be achieved for EV-DO RevA by a 40-slot buffer because there are eight slots between each four-slot subpacket). have. Undecoded subpackets appear dark. Decoded subpackets are dim and discarded in the 40-slot buffer. Each time instance corresponds to the arrival of another subpacket on the interlace. At slot time n, four stored subpackets of User 1 are correctly decoded, while the most recent subpackets from Users 2 and 3 fail to decode.

In time instance n + 12 slots, consecutive subpackets of the interlace arrive with interference cancellation of user 1's decoded (non-dark) subpackets 2, 3 and 4. During time instance n + 12 slots, packets from users 2 and 3 are successfully decoded.

10 applies the IC to a group of users on the same frame offset but does not perform continuous interference cancellation within the group. In a typical group IC, users of the same group are not aware of mutual interference cancellation. Therefore, the implementation complexity decreases as the number of users in the group increases, but there is a loss due to lack of deletion between users of the same group for the same decoding attempt. However, with H-ARQ, the receiver attempts to decode all users in the group after each new subpacket arrives, allowing users in the same group to achieve mutual interference cancellation. For example, when User 1's packet decodes at time n, this helps the User 2 and User 3's packets decode at time n + 12 and also helps User 1 decode at time n + 24. All subpackets of a previously decoded packet may be discarded before retrying decoding on them when the next subpackets of other users arrive. The important point is that certain users may always be in the same group, but their subpackets are aware of the IC gain when other group members decode. Eliminate joint interference in pilot, overhead, and traffic channels

The problem addressed by this section relates to improving the system capacity of the CDMA RL by efficiently estimating and eliminating multi-user interference at the base station receiver. In general, the RL user's signal consists of pilot, overhead, and traffic channels. This section describes the joint pilot, overhead, and traffic IC schemes for all users.

There are two forms described. First, an overhead IC (OIC) is derived. On the reverse link, the overhead from each user acts as an interference to all other user's signals. For each user, the collective interference due to overhead by all other users may be a large percentage of the total interference experienced by that user. The removal of this aggregate overhead interference may be (e.g., CDMA2000 ^® 1xEV-DO Rev for the system) further improve the system performance and increase reverse link capacity beyond performance and capacity achieved by PIC and TIC Can be.

Second, significant interactions between PICs, OICs, and TICs are demonstrated through system performance and hardware (HW) design transactions. Several ways of how best to combine all three removal procedures are described. Some schemes may have greater performance gains, and some schemes may have more complexity advantages. For example, one of the approaches described removes all pilot signals before decoding any overhead and traffic channels, and then decodes and removes the overhead and traffic channels of the user in a sequential manner.

This section is based on the CDMA2000 ^® 1x EV-DO Rev system, and generally applies to other CDMA systems such as W-CDMA, CDMA2000 ^® 1x, and CDMA2000 ^® 1x EV-DV. How to delete an overhead channel

11 shows an RL overhead channel structure as for EV-DO RevA. There are two types of overhead channels: one type is to assist the RL demodulation / decoding, which includes a reverse rate indicator (RRI) channel and an auxiliary pilot channel (used when the payload size is 3072 bits or greater). Another type is to use a forward link (FL) function that includes a data rate control (DRC) channel, a data source control (DSC) channel, and an acknowledgment (ACK) channel. As shown in FIG. 11, ACK and DSC channels are time multiplexed on a slot basis. The ACK channel is sent only upon acknowledgment of a packet sent to the same user on the FL.

식 1 재구성된 파일럿 신호들 여기서 n은 chipx1 샘플링 레이트에 해당하고, f는 핑거 번호이며, c _f 는 PN 시퀀스이고, w _{f , aux} 는 보조 파일럿 채널에 할당되는 왈시 코드이며, G _aux 는 기본 파일럿 채널에 대한 이 채널의 상대 이득이고, h _f 는 하나의 세그먼트에 대해 상수인 것으로 가정되는 추정된 채널 계수(또는 채널 응답)이며, φ는 송신 펄스와 chipx8 해상도 의 수신기 저역 필터의 필터 함수 또는 컨볼루션이고(φ는 [-MT _c , MT _c ]에서 사소하지 않은 것으로 추정됨), γ _f 는 α _f = γ _f mod 4 및

에 의한 이 핑거의 chipx8 시간 오프셋이다.Among the overhead channels, the data of the auxiliary pilot channel is deductively known to the receiver. Thus, similar to the primary pilot channel, this channel does not require demodulation and decoding, and the secondary pilot channel can be reconstructed based on information about the channel. The reconstructed secondary pilot can be 2x chip rate in resolution and can be expressed as follows (for one segment):

Equation 1 Reconstructed Pilot Signals where n corresponds to the chipx1 sampling rate, f is a finger number, c _f is a PN sequence, w _{f , aux} is a Walsh code assigned to the auxiliary pilot channel, and G _aux is the primary pilot The relative gain of this channel over the channel, h _f is the estimated channel coefficient (or channel response) that is assumed to be constant for one segment, and φ is the filter function or convolution of the receiver low pass filter with transmit pulse and chipx8 resolution. And φ is not trivial in [ -MT _c , MT _c ], γ _f is α _f = γ _f mod 4 and

This is the chipx8 time offset of this finger.

식 2 재구성된 오버헤드(DRC, DSC, RRI) 신호들The second group of overhead channels, including the DRC, DSC and RRI channels, are encoded by binary orthogonal code or simple code. On the receiver side, for each channel the first demodulated output is compared with a threshold. If the output is less than the threshold, an erase is declared and no reconstruction is attempted for this signal. Otherwise, the outputs are decoded by a symbol based maximum likelihood (ML) detector, which can be in decoder 308 of FIG. 4. The decoded output bits are used for reconstruction of the channel as shown in FIG. The reconstructed signals for these channels are given as follows:

Equation 2 Reconstructed Overhead (DRC, DSC, RRI) Signals

Compared with Equation 1, there is one new term d ₀ , which is overhead channel data, w _{f , 0} is Walsh cover, and G _aux represents the overhead channel gain for the basic pilot.

The remaining overhead channel is a 1 bit ACK channel. This can be BPSK modulation, uncoding and repeated over half a slot. The receiver can demodulate the signal and make a hard decision on the ACK channel data. The reconstruction signal model may be as in Equation 2.

식 3 그리고 재구성을 위해, 전송된 비트의 소프트 추정치는 다음과 같을 수 있다:

식 4 여기서 tanh 함수는 도표 작성될 수 있다. 재구성된 ACK 신호는 d ₀ 를

로 대체한 것을 제외하고 도 2와 매우 비슷하다. 일반적으로, 수신기는 데이터를 확실히 모르고 이 방법은 신뢰 레벨을 그림으로 제시하기 때문에 소프트 추정 및 제거 접근은 더 나은 제거 성능을 가져야 한다. 이러한 접근은 일반적으로 상술한 오버헤드 채널들로 확장될 수 있다. 그러나 각 비트에 대한 LLR을 얻기 위한 최대 사후 확률(MAP) 검출기의 복잡도는 한 코드 심벌의 정보 비트 수에 따라 지수적으로 증가 한다.Another approach to reconstruct the ACK channel signal assumes that the demodulated and accumulated ACK signal can be expressed after normalization as follows: y = x + z where x is a transmitted signal and z is a scaled variance of σ ² . Noise term. And the likelihood ratio (LLR) of y is given by:

For Equation 3 and reconstruction, the soft estimate of the transmitted bits may be as follows:

Where the tanh function can be plotted. The reconstructed ACK signal is _equal to d ₀

It is very similar to FIG. In general, the soft estimation and rejection approach should have better rejection performance because the receiver does not know the data clearly and this method presents the confidence level graphically. This approach can generally be extended to the overhead channels described above. However, the complexity of the maximum posterior probability (MAP) detector for obtaining the LLR for each bit increases exponentially with the number of information bits of one code symbol.

는

가 된다. 식 5 조인트 PIC , OIC , TIC One efficient way to implement overhead channel reconstruction is that one finger can scale each decoded overhead signal to its relative gain, cover it with a Walsh code, sum it, and then spread it into one PN sequence And both can be filtered simultaneously through the channel scale filter hφ . This method reduces both memory bandwidth and computational complexity for deduction.

Is

Becomes Equation 5 Joints PIC , OIC , TIC

Joints PIC, OIC, and TIC can be performed to achieve high performance and increase system capacity. Different decoding and erasing sequences of PIC, OIC, and TIC can yield different effects on hardware design complexity and different system performance. PIC first, then OIC and TIC together (first method)

12A illustrates a method of first performing PIC and then performing OIC and TIC together. At start block 1200, the receiver derives channel estimates for all users and performs power control at block 1202. Since pilot data for all users is known to the BTS, they can be subtracted if their channels are estimated at PIC block 1204. Thus, all user traffic channels and certain overhead channels can observe less interference and benefit from forward pilot rejection.

Block 1206 selects a group G of undecoded users, eg, users whose packets or subpackets end at the current slot boundary. Blocks 1208-1210 perform overhead / traffic channel demodulation and decoding. At block 1212, only successfully decoded channel data will be reconstructed and excluded from the FERAM 312 shared by all users. Block 1214 checks if there are more users to be decoded. Block 1216 ends the process.

Decoding / reconstruction / deletion may be a sequential manner from one user in the group to the next user in the group, which may be referred to as continuous interference cancellation. In this approach, users of the latter decoding order of the same group benefit from the deletion of users of the faster decoding order. The simplified approach is to decode all users in the same group and then subtract all of their interference contributions at the same time. The second approach or approach (described below) enables both lower memory bandwidth and more efficient pipeline structures. In both cases, packets of users that do not end at the same slot boundary but overlap with this group of packets benefit from this deletion. Such deletion can account for most of the deletion gains in asynchronous CDMA systems.

12B illustrates an apparatus that includes means 1230-1244 to perform the method of FIG. 12A. The means 1230-1244 of FIG. 12B may be implemented in hardware, software, or a combination of hardware and software.

13A illustrates a variation of the method of FIG. 12A. Blocks 1204-1210 remove the signal based on the initial channel estimate at block 1202. Block 1300 derives a data based channel estimate or a refined channel estimate. The data-based channel estimate may provide a better channel estimate as described below. Block 1302 performs the remaining PIC, that is, removes the calibrated estimate of the signal based on the channel estimate refinement at block 1300.

For example, assume that the original signal estimate (eg, pilot signal) P1 [n] has been removed from the samples received at blocks 1204-1210. Then, based on the better channel estimate derived at block 1300, the method forms a calibrated signal estimate P2 [n]. The method may remove a progressive P2 [n] -P1 [n] difference from the sample location in RAM 312.

FIG. 13B illustrates an apparatus including means 1230-1244, 1310, 1312 to perform the method of FIG. 13A. The means 1230-1244, 1310, 1312 of FIG. 13B may be implemented in hardware, software, or a combination of hardware and software. First PIC , then OIC , then TIC (second method)

This second approach is similar to FIG. 12A described above, except that overhead channels of the same user group are demodulated and decoded before any traffic channels are demodulated and decoded. This approach is suitable for non-interlaced systems because no strict ACK deadline is imposed. Interlaced systems, for example, DO Rev. In the case of A, since the ACK / NAK signal corresponds to a traffic channel subpacket, the allowable decoding delay for the traffic channel subpacket is generally limited to within a pair of slots (1 slot = 1.67 ms). Thus, if some overhead channels spread over more than this time scale, this approach may not be feasible. In particular, in DO RevA, the auxiliary pilot channel and the ACK channel are short duration formats and can be excluded before the TIC. Joint Pilot / Overhead / Traffic Channel Drop (Third Method)

14A shows a method of performing joints PIC, OIC and TIC. After the start block 1400, at block 1402 the receiver derives channel estimates for all users and performs power control. Block 1404 selects a group G of undecoded users. Block 1406 reestimates the channel from the pilot. Blocks 1408-1410 attempt to perform overhead / traffic channel demodulation and decoding. Block 1412 performs OIC and TIC only for PICs for all users and only for successfully decoded users.

The pilot differs from FERAM 312 immediately after channel estimation for all users (block 1402) and differs from the first scheme (FIG. 12a) in that channel estimates are used for power control, such as non-IC schemes. The method then performs sequential decoding (blocks 1408 and 1410) in a predetermined order, for groups of users that ended at the same packet / sub packet boundary.

For the attempted decoding user, the method first reestimates the channel from the pilot (block 1402). The pilot identifies less interference compared to the demodulated time for power control due to interference cancellation of previously decoded packets overlapping the traffic packet to be decoded (block 1402). Therefore, channel estimation quality is improved, which is advantageous for both traffic channel decoding and cancellation performance. This new channel estimation is used not only for traffic channel decoding (block 1410) but also for overhead channel decoding (block 1408) (eg, RRI channel in EV-DO). Upon completion of the decoding process for a user at block 1412, the method will subtract this user's interference contribution from the FERAM 312, which includes a pilot channel and any decoded overhead / traffic channel.

Block 1414 checks if there are more users to decode. Block 1416 ends the process.

14B illustrates an apparatus that includes means 1420-1436 to perform the method of FIG. 14A. The means 1420-1436 of FIG. 14B may be implemented in hardware, software, or a combination of hardware and software.

15A illustrates a variation of the method of FIG. 14A. Block 1500 derives the data based channel estimate. Block 1502 performs an optional remaining PIC as in FIG. 13A.

FIG. 15B shows an apparatus comprising means 1420-1436, 1510, 1512 to perform the method of FIG. 15A. The means 1420-1436, 1510, 1512 of FIG. 15B may be implemented in hardware, software, or a combination of hardware and software. Transaction between the first scheme and the third scheme

It may appear that the first scheme should have better performance than the third scheme since the pilot signals are known to the BTS and detect their deletion in the front. If both schemes are assumed to have the same erasure quality, the first scheme can outperform the third scheme over all data rates. However, in the first scheme, since the pilot channel estimation confirms higher interference than traffic data demodulation, the estimated channel coefficients used for reconstruction (for both pilot and overhead / traffic) may be more noisy. However, in the third scheme, since the pilot channel estimation is performed immediately after the traffic data demodulation / decoding, the interference level confirmed by this refined channel estimation is the same as the traffic data demodulation. And on average, the erase quality of the third scheme may be better than the first scheme.

From a hardware design perspective, the third scheme can have some edges, which can sum up pilot and decoded overhead and traffic channel data and delete them together, thus this approach can save memory bandwidth. . On the other hand, re-estimation of the pilot can be performed with overhead channel demodulation or traffic channel demodulation (with respect to sample reading from memory), so there is no increase in memory bandwidth requirements.

Assuming that the first scheme has 80% or 90% deletion quality of the third scheme, there is a trade-off between data rate per user versus gain for the number of users. In general, the first approach is advantageous if all the users are in the low data rate range, and vice versa if all are high data rate users. The method may also reestimate the channel from the traffic channel once one data packet is decoded. Since the traffic channel operates at (much) higher SNR than the pilot channel, the drop quality is improved.

Overhead channels may be removed (deleted) upon successful demodulation and traffic channels may be removed when successfully demodulated and decoded. It is possible for a base station to successfully demodulate / decode the overhead and traffic channels of all access terminals at the same time. If this occurs (PIC, OIC, TIC), the FERAM will contain only the remaining interference and noise. Pilot, overhead, and traffic channel data may be deleted in various orders, and may be deleted for subsets of access terminals.

One approach is to perform interference cancellation (of any combination of PIC, TIC, OIC) for one user from RAM 312 at the same time. Another approach is to (a) accumulate the reconstructed signal (of any combination of PIC, TIC, OIC) for a group of users, and (b) perform interference cancellation on that group simultaneously. These two approaches can be applied to any of the methods, methods, and processes disclosed herein. Improvement of Channel Estimation for Interference Cancellation

The ability to accurately reconstruct received samples can have a significant impact on the system performance of a CDMA receiver that implements interference cancellation by reconstructing and removing various components of the transmitted data. In a rake receiver, the multipath channel is estimated by PN despreading on the pilot sequence and then pilot filtering (ie, cumulative) over the appropriate period. The length of the pilot filtering is typically chosen as a compromise between increasing the estimated SNR by accumulating more samples and not accumulating long enough for the estimated SNR to deteriorate by the time change of the channel. The channel estimate from the pilot filter output is used for data demodulation.

As described above with respect to FIG. 4, one practical way to implement interference cancellation in a CDMA receiver is to reconstruct the contribution of the various transmitted chipx1 streams to (eg chipx2) FERAM samples. This involves determining an estimate of the overall channel and the chip stream transmitted between transmitter chips and receiver samples. Since the channel estimates from the rake fingers represent the multipath channel itself, the overall channel estimate should also take into account the presence of transmitter and receiver filtering.

This section discloses several ways to improve this overall channel estimation for interference cancellation in a CDMA receiver. These technologies can be applied to CDMA2000 ^® , 1xEV-DO, 1xEV-DV, and WCDMA.

To perform TIC of the packet to decode correctly, the receiver of FIG. 4 takes the information bits from the decoder output and reconstructs the transmitted chip stream by re-encoding, reinterleaving, remodulation, reapplying and respreading the data channel gains. can do. To estimate the received samples for the TIC by the pilot channel estimate, the transmit chip stream is wound with a rake receiver channel estimate from the model of the transmitter and receiver filters and the despreading by the pilot PN sequence.

Instead of using the pilot channel estimate, an improved channel estimate (at each Rake finger delay) can be obtained by despreading by the reconstructed data chips themselves. This improved channel estimate is not useful for data demodulation of the packet because the packet has already been correctly decoded but is only used to reconstruct the contribution of this packet to the frontend samples. By this technique, for each delay of the Rake fingers (eg, chipx8 resolution), the method “despreads” the received samples (eg, interpolated on chipx8) into a reconstructed data chip stream, and Can accumulate over time. This improves channel estimation because the traffic channel is transmitted at a higher power than the pilot channel (this traffic-pilot T2P ratio is a function of data rate). Using data chips for channel estimation for TIC can make channel estimation more accurate for higher power users, which is most important for high accuracy cancellation.

Instead of estimating the multipath channel at each Rake finger delay, this section describes a channel estimation procedure that explicitly estimates the combined effects of the transmitter filter, the multipath channel and the receiver filter. This estimation may be the same resolution as the oversampled frontend samples (eg chipx2 FERAM). Channel estimation may be accomplished by despreading frontend samples into reconstructed transmit data chips to achieve T2P gain in channel estimation accuracy. The time range of evenly spaced channel estimates may be selected based on information about the Rake finger delay and a deductive estimate of the composite response of the transmitter and receiver filters. Moreover, the information from the rake fingers can be used to refine the evenly spaced channel estimates.

식 6 여기서 복소 경로 진폭은 해당 지연(τ _l )을 갖는 a _l 이다. 송신기 및 수신기 필터의 복합적인 영향은 다음과 같이 φ(t)로 정의될 수 있으며:

식 7 여기서

는 컨볼루션을 나타낸다. 합성된 φ(t)는 종종 상승 코사인 응답과 비슷하게 선택된다. 예를 들어, CDMA2000^® 및 그 도함수에서, 응답은 도 17에 디스플레이되는 예 φ(t)와 비슷하다. 전체 채널 추정치는 다음과 같이 주어진다:

식 816 shows a model of a transmission system with a transmission filter p (t), a full / synthetic channel h (t) (for the multipath channel g (t) described below), and a receiver filter q (t). The digital baseband representation of a wireless communication channel can be modeled with L discrete multipath components as follows:

Where the complex path amplitude is a _l with the corresponding delay τ _l . The combined effect of the transmitter and receiver filters can be defined as φ (t) as follows:

Equation 7 where

Indicates convolution. The synthesized φ (t) is often chosen to be similar to the synergistic cosine response. For example, CDMA2000 ^® and in that derivatives, the response is similar to the example φ (t) displayed in FIG. The overall channel estimate is given by:

Equation 8

18A and 18B show examples of channel estimates (real and imaginary components) based on estimated multipath channels at each of the three rake fingers. In this example, the actual channel is represented by a solid line and a _l is given by an asterisk. Reconstruction (dotted line) is based on the use of a _l in Equation 6 above. The Rake finger channel estimates in Figures 18A and 18B are based on despreading by the pilot chip (the overall pilot SNR is -24 dB). Rake by Regenerated Data Chips Instead of Pilot Chips Despreading of the finger delays

Channel estimation quality has a direct impact on fidelity, which reconstructs the user's contribution to the received signal. In order to improve the performance of a CDMA system that implements interference cancellation, the user's reconstructed data chips can be used to determine an improved channel estimate. This will improve the accuracy of the interference deduction. One technique for CDMA systems may be described as "despreading for the user's transmitted data chips" as opposed to conventional "despreading for the user's transmitted pilot chips."

Recall that the Rake finger channel estimates in Figures 18A and 18B are based on despreading by pilot chips (where the total pilot SNR is -24 dB). 19A and 19B show an example of an improved channel estimate based on despreading by rake fingers and data chips, where the data chips are transmitted at 10Os higher power than pilot chips.

20A shows a despreading method in Rake finger delay with regenerated data chips. At block 2000, rake receiver 314 (FIG. 4) despreads frontend samples with pilot PN chips to obtain rake finger values. In block 2002, demodulator 304 performs data demodulation. In block 2004, the decoder 308 performs data decoding and checks the CRC. At block 2006, when passing the CRC, unit 400 determines the data chips transmitted by reencoding, reinterleaving, remodulation, and respreading. In block 2008, unit 400 despreads frontend samples with the transmitted data chips to obtain an improved channel estimate at each finger delay. At block 2010, unit 400 reconstructs the user's traffic and overhead contribution to frontend samples with improved channel estimates.

FIG. 20B shows an apparatus comprising means 2020-2030 to perform the method of FIG. 20A. The means 2020-2030 of FIG. 20B may be implemented in hardware, software, or a combination of hardware and software. Composite Channel Estimation of FERAM Resolution by Regenerated Data Chips

Conventional CDMA receivers can estimate the complex value of a multipath channel at each rake finger delay. The receiver frontend in front of the rake receiver includes a low pass receiver filter (ie q (t)) that matches the transmitter filter (ie p (t)). Thus, for a receiver to implement a filter that matches the channel output, the rake receiver itself attempts to match only the multipath channel (i.e. g (t)). The delay of the rake fingers is typically derived from an independent time tracking loop within the minimum separation requirement (eg, fingers are at least one chip apart). However, the physical multichannel itself can often have energy in successive delays. Thus, some methods estimate the composite channel (i.e., h (t)) at the resolution of the frontend samples (e.g., chipx2 FERAM).

With transmit power control on the CDMA reverse link, the synthesized finger SNRs from the multipath and receiver antennas are typically controlled to be within a certain range. This range of SNRs can cause the composite channel estimates to be derived from despread pilot chips with relatively large estimate deviations. This is because the rake receiver only wants to place the fingers at the "peak" of the energy delay profile. However, with the T2P advantage of despreading by the reconstructed data chips, the composite channel estimate can obtain a better h (t) estimate than the direct estimate of the combined φ (t) model and g (t).

The channel estimation procedure described herein explicitly estimates the combined effects of the transmitter filter, the multichannel and the receiver filter. This estimation may be the same resolution as the oversampled frontend samples (eg chipx2 FERAM). Channel estimation may be accomplished by despreading frontend samples into reconstructed transmit data chips to achieve T2P gain in channel estimation accuracy. The time range of evenly spaced channel estimates may be selected based on information about the Rake finger delay and a deductive estimate of the composite response of the transmitter and receiver filters. Moreover, the information from the rake fingers can be used to refine the evenly spaced channel estimates. Note that the technique of estimating the synthesis channel itself is useful because it does not require the design to use the deductive estimate of φ (t).

21A and 21B show an example of estimating a composite channel using evenly spaced samples at chipx2 resolution. In FIGS. 21A and 21B, the data chip SNR is -4 dB, which corresponds to a pilot SNR of -24 dB and a T2P of 20 dB. Uniform channel estimates provide better quality compared to despreading of only the data chips at the rake finger position. At high SNR, the impact of the "fat path" limits the ability to accurately reconstruct the channel using the Rake finger position. The uniform sampling approach is particularly useful when the estimated SNR is high, which is the case of despreading with data chips for high T2P. Channel reconstruction fidelity is important when the T2P for a particular user is high.

22A illustrates a method of estimating a composite channel at a constant resolution using regenerated data chips. Blocks 2000-2006, 2010 are similar to FIG. 20A described above. At block 2200, rake receiver 314 (FIG. 4) or other component determines a time range for even configuration based on the rake finger delay. At block 2202, demodulator 304 determines an improved channel estimate by despreading frontend samples with data chips transmitted with an even delay for an appropriate time range.

FIG. 22B shows an apparatus comprising means 2020-2030, 2220, 2222 to perform the method of FIG. 22A. The means 2020-2030 of FIG. 22B may be implemented in hardware, software, or a combination of hardware and software.

In the above description, g (t) is the wireless multipath channel itself, and h (t) includes the wireless multipath channel as well as transmitter and receiver filtering: g (t) wound with h (t) = phi (t). .

In the above description, the "sample" can be any rate (eg, twice per chip), while the "data chip" is one per chip.

"Regenerated data chips" are formed in block 2006 of FIG. 20A and formed by re-encoding, reinterleaving, remodulation, and re-spreading as described above. In principle, "regeneration" mimics the process by which information bits pass through a mobile transmitter (access terminal).

“Reconstructed Samples” refers to samples stored in separate memory from FERAM 312 or FERAM 312 of the receiver (eg, twice per chip). These reconstructed samples are formed by winding the (regenerated) transmitted data chips with a channel estimate.

The terms "reconstructed" and "regenerated" may be used interchangeably when provided for calibration of transmitted data chips or calibration of received samples. The "chips" are calibrated by re-encoding, etc., while the "samples" are calibrated based on the use of the calibrated chips and the integration of the effects of the radio channel (channel estimate) and transmitter and receiver filtering. Can be. The terms "reconstruction" and "regeneration" both mean original reconstruction or calibration. There is no technical distinction. Some embodiments exclusively use "regeneration" for data chips and "reconstruction" for samples. And the receiver may have a data chip regeneration unit and a sample reconstruction unit. Adaptation of Transmit Subchannel Gain on Reverse Link of CDMA Systems by Interference Cancellation

Multi-user interference is a finite factor in CDMA transmission systems, and any receiver technology that mitigates this interference may enable significant improvements in throughput that can be achieved. This section describes techniques for adaptation of the transmit subchannel gain of the system by the IC.

In reverse link transmission, each user transmits pilot, overhead, and traffic signals. Pilots provide synchronization and estimation of the transmission channel. Overhead subchannels (such as RRI, DRC, DSC, ACK) are needed for MAC and traffic decoding setup. Pilot, overhead, and traffic subchannels have different requirements for SINR. In a CDMA system, a single power control can adapt the pilot's transmit power, while the overhead and traffic subchannel's power has a constant gain for the pilot. When the BTS is equipped with PIC, OIC, and TIC, the various subchannels identify different levels of interference depending on the order and erasure capability of the IC. In this case, a static relationship between subchannel gains may impair system performance.

This section describes the new gain control power for the different logical subchannels on the system implementing the IC. Technique may be applied to, and based on CDMA systems such as EV-DO RevA, EV-DV Rel D, W-CDMA EUL , and CDMA2000 ^®.

The techniques described implement power and gain control for different subchannels by adaptively changing the gain of each subchannel according to the measured performance in terms of packet error rate, SINR or interference power. The goal is to provide a reliable power and gain control mechanism that can take full advantage of the IC's potential while providing strength for transmission over time-varying distributed subchannels.

Interference cancellation involves removing the contribution of logical subchannels to frontend samples after the logical subchannels are decoded to reduce interference to other signals to be decoded later. In the PIC, the transmitted pilot signal is known to the BTS and the received pilot is reconstructed using the channel estimate. In the TIC or OIC, the interference is eliminated by reconstructing the received subchannel through the version decoded in the BTS.

The current BTS (without IC) controls the power E _cp of the pilot subchannel to meet the error rate requirement in the traffic channel. The power of the traffic subchannel is related to the pilot by a constant factor T2P, which depends on the payload type and target termination target. The adaptation of pilot power is performed by a closed loop power control mechanism that includes an inner and an outer loop. The inner loop aims to keep the pilot's SINR (E _cp / Nt) at the threshold level T, while the outer loop power control changes the threshold level T based on, for example, the packet error rate (PER). do.

When IC is performed at the receiver (Figure 4), adaptation of the subchannel gain may be beneficial to the system. In fact, since each subchannel identifies a different level of interference, the gain for the pilots must be adapted accordingly to provide the desired performance. This section addresses the challenges of gain control over overhead and pilot channels, and describes techniques for adapting T2P to fully increase the throughput of the system by fully utilizing the IC. Important parameters in the system by IC

Two parameters that can be adjusted are the overhead subchannel gain and the traffic to pilot (T2P) gain. When TIC is enabled, the overhead subchannel gain is increased (relative to non-TIC) to enable more flexible trade between pilot and overhead performance. By representing the base line (G) used in the current system, the new value of the overhead channel gain will be: G ' = G · Δ _G Equation 9

In a non-IC scheme, overhead / pilot subchannels identify the same interference level as traffic channels, and certain non-T2P / Gs can provide satisfactory performance for both overhead and traffic channel performance as well as pilot channel estimation. have. If an IC is used, the interference levels for overhead / pilot and traffic are different and T2P can be reduced to enable coherent performance of both types of subchannels. For a given payload, the method may reduce the factor Δ _{T 2 P} by T2P on Table value to satisfy the requirements. By representing the base line (T2P) used for a particular payload in the current system as T2P, the new value of T2P would be: T 2 P ' = T 2 P · Δ _{T 2 P} Equation 10

The parameter Δ _{T 2 P} may be quantized into a set of finite or discrete values (eg, −0.1 μs to −1.0 μs) and sent to the access terminal 106.

Some quantities that can be kept under control are traffic PER, pilot SINR, and rise over thermal (ROT) DLEK. The pilot SINR should not fall below the minimum level desired for good channel estimation. ROT is important to ensure the stability and link budget of power controlled CDMA reverse links. In a non-TIC receiver, ROT is defined for the received signal. In general, the ROT should be kept within a predetermined range to enable good capacity / coverage transactions. Rise Over Thermal (ROT) Control

식 11 I ₀ ' 값은 IC에 의해 업데이트된 후 프론트엔드 샘플들로부터 측정될 수 있다. IC가 수행될 때 ROT는 오버헤드 서브 채널에 여전히 중요하고, ROT는 임계치에 대해, 즉 다음을 만족하도록 제어되어야 한다:

식 12 여기서 N ₀ 은 잡음 전력이다. I ₀ represents the signal power at the input of the receiver. Interference cancellation from the received signal results in power reduction. I ₀ ′ represents the average power of the signal at the input of demodulator 304 after IC:

Equation 11 I ₀ 'value can be measured from frontend samples after being updated by the IC. The ROT is still important for the overhead subchannels when the IC is performed, and the ROT must be controlled for a threshold, i.e. to satisfy the following:

Where N ₀ is the noise power.

식 13 잡음 레벨이 변하지 않는다는 가정 하에서 ROT _eff 에 대한 제약은 I ₀ '에 대한 제약으로서 동등하게 식으로 나타낼 수 있다:

I ₀ ^(thr) 는 ROT _thr ^{( eff )} 에 대응하는 신호 전력 임계치이다. 일정한 오버헤드 이득 기술 However, traffic and some overhead subchannels benefit from the IC. The decoding performance of the subchannels is related to the ROT measured after the IC. The effective ROT is the ratio of the signal power after the IC to the noise power. The effective ROT can be controlled by a threshold, ie:

Equation 13 Assuming that the noise level does not change, the constraint on ROT _eff can be expressed equally as the constraint on I ₀ ' :

I ₀ ^(thr) is the signal power threshold corresponding to ROT _thr ^{( eff )} . Constant Overhead Gain Technology

As the ROT increases, the SINR of the pilot and overhead channels (which do not benefit from the IC) decreases, potentially increasing the error rate. To compensate for this effect, the overhead channel gain may be raised by a constant value or by adaptation to specific system conditions.

A technique is described in which the gain of an overhead subchannel is fixed for a pilot. The proposed technique is suitable for both the level of pilot subchannel and Δ _{T 2 P} for each user. Closed-loop control of T2P with constant Δ _G = 0 ㏈

FIG. 23 shows closed loop power control (PC) for E _cp and Δ _{T 2 P} and constant Δ _G = 0 μs (block 2308). This first solution for the adaptation of E _cp and Δ _{T 2 P} includes:

A. Inner and outer loops 2300 and 2302 may perform power control in a conventional manner for adaptation of E _cp . Outer loop 2300 receives target PER and traffic PER. Inner loop 2304 receives threshold T 2302 and the measured pilot SINR and outputs E _cp .

B. Closed Loop Gain Control (GC) 2306 adapts _{ΔT 2 P} based on the measurement of the removed interference. Gain control 2306 receives the measured ROT and the measured ROT _eff and outputs _{ΔT 2 P.} The receiver measures the interference removed by the IC scheme and adapts Δ _{T 2 P.}

C. Δ _{T 2 P} may be sent in a message periodically to all access terminals 106 in the sector.

식 14Δ _T for adaptation of _{2 P,} if the interference after IC reduced to I ₀ 'from I _0, as a result T2P may be reduced as follows:

Equation 14

식 15E _cp will increase (through PC loop 2304 as follows:

Equation 15

식 16 여기서 G는 오버헤드 채널 이득이다. (G에 대한) 가장 큰 T2P 값의 경우, 비(C)는 다음과 같이 근사화될 수 있다:

식 17The ratio between total transmit power for a system with an IC and a system without an IC will be:

Where G is the overhead channel gain. For the largest T2P value (relative to G), the ratio C can be approximated as follows:

Equation 17

To estimate the effective ROT, the effective ROT changes rapidly due to changes in PC and channel conditions. Instead, Δ _{T 2 P} reflects the slow variation of ROT _eff . Therefore, for the selection of _{ΔT 2 P} , the effective ROT is measured by the long average window of the signal behind the IC. The average window may have at least twice the length of the power control update period. Constant Δ _G> 0 ㏈ the closed-loop control of T2P

식 18 Δ _{T 2 P} 의 ACK 기반 제어 24 is a gain control 2306 receives a threshold effective ROT, and is similar to the Δ _G> 0 ㏈ and 23 except that the (block 2400). This alternative method for adaptation of _{ΔT 2 P} is based on the requirement that both IC and non-IC systems have the same cell coverage. E _cp distribution is the same in both cases. The effect of the IC is double for a fully loaded system: i) the signal power I ₀ before the IC will increase compared to the signal power of a system without the IC; Ii) Due to closed loop power control by PER control, I ₀ ′ will be similar to the signal power of a system without IC. Δ _{T 2 P} is adapted as follows:

ACK- based control of equation 18 Δ _{T 2 P}

FIG. 25 shows the PC for E _cp and Δ _{T 2 P} based on the ACK subchannel with constant overhead subchannel gain (block 2506).

The closed loop GC of Δ _{T 2 P} requires a feedback signal from the BTS to the AT, and all ATs receive the same broadcast value of Δ _{T 2 P} from the BTS. An alternative solution is based on open loop GC 2510 and closed loop PCs 2500 and 2504 for Δ _{T 2 P.} Closed loop pilot PC includes an inner loop 2504, which adjusts E _cp according to threshold T ₀ 2502. Outer loop control 2500 is indicated by the error rate of the overhead subchannels, for example, data rate control (DRC) subchannel error probability or DRC error rate. T ₀ increases each time the DRC error rate exceeds the threshold, but gradually decreases when the DRC error rate is below the threshold.

Δ _{T 2 P} is adapted over the ACK forward channel. In particular, by measuring the statistics of ACK and NACK, the AT may evaluate the traffic PER at the BTS (block 2508). Gain control 2510 compares the target traffic PER and the measured PER. Each time the PER is above the threshold, Δ _{T 2 P} increases until T2P 'reaches the baseline value (T2P) of the non-IC system. On the other hand, for a lower PER, Δ _{T 2 P} decreases to fully utilize the IC process. Variable Overhead Gain Technology

Further optimization of the transceiver can be obtained by adapting the _{ΔT 2 P} as well as the overhead subchannel gain (G overhead) to the IC process. In this case, an extra feedback signal is needed. The value of Δ _G can be quantized to 0 ㏈ to 0.5 ㏈. Interference Power Based Overhead Gain Control

식 19FIG. 26 is similar to FIG. 24 except for overhead GC 2600. Overhead sub-channel 2600 is based on the signal power measured behind the IC. In this case, E _cp is estimated to provide the same cell coverage of the system without IC. The signal before the IC has increased power I ₀ and the overhead gain compensates for the increased interference. This implementation adapts the overhead gain by setting:

Equation 19

[Delta] _G can be controlled to not go below zero [mu] s, since it corresponds to a reduction in overhead subchannel power that is not readily available.

Gain and power control schemes include the inner and outer loop PCs 2304 and 2300 for E _cp as in FIG. 23, the GC loop 2600 for Δ _G as described above, and the open loop GC for Δ _{T 2 P} ( 2306), where _{ΔT 2 P} increases each time the PER is above the target value and decreases when the PER is below the target. The maximum level of Δ _{T 2 P} is allowed, which corresponds to the level of the non-IC receiver. DRC-only overhead gain control

FIG. 27 illustrates a variation of FIG. 26 by DRC dedicated overhead gain control 2702.

Even when the overhead subchannel gain is adapted, the gain control 2700 of _{ΔT 2 P} may be performed in a closed loop as described above. In this case, E _cp and Δ _{T 2 P} are controlled as in the scheme of FIG. 23, but adaptation of the overhead subchannel gain 2702 is performed via DRC error rate. In particular, if the DRC error is above the threshold, the overhead subchannel gain 2702 is increased. Overhead gain 2702 decreases gradually when the DRC error rate is below a threshold. T2P Control in Multisector Multicell Networks

식 20 여기서 Δ _{T 2 P} ^(s) 는 섹터(s)에 의해 요청된 Δ _{T 2 P} 이다. AT는 다양한 셀로부터 서로 다른 요청을 수신할 수 있으며, 또한 이 경우 다양한 기준이 채택될 수 있다. 방법은 서빙 섹터와의 가장 신뢰성 있는 통신을 보장하기 위해 서빙 섹터에 대응하는 Δ _{T 2 P} 를 선택할 수 있다.Since the GC of Δ _{T 2 P} may be performed at the cell level and the AT 106 may be in softer handoff, the various sectors may generate different adaptation requests. In this case, various options may be considered for the selection of the _{ΔT 2 P} request to be sent to the AT. At the cell level, the method may select the minimum reduction in T2P among those requested by fully loaded sectors, ie

Equation 20 wherein Δ _{T 2 P} ^(s) is Δ _{T 2 P} requested by sector (s). The AT may receive different requests from various cells, in which case various criteria may be adopted. The method may select _{ΔT 2 P} corresponding to the serving sector to ensure the most reliable communication with the serving sector.

For the selection of _{ΔT 2 P} in the cell and AT, other selections including minimum, maximum or average of the requested values may be considered.

One important form is for the mobile station to use T2P '= T2P × Δ _{T 2 P} and G' = G × Δ _G , where Δ _{T 2 P} is a measure of I ₀ and I ₀ ′ (and possibly I ₀ ^thr as well). ) was calculated from the BTS to the base, Δ _G is also calculated at the BTS. With these delta_factors computed in the BTS, they are broadcast by every BTS to all access terminals, and the access terminals respond accordingly.

The concept disclosed herein can be applied to a WCDMA system using overhead channels such as a dedicated physical control channel (DPCCH), an extended dedicated physical control channel (E-DPCCH) or a high speed dedicated physical control channel (HS-DPCCH). The WCDMA system may use a dedicated physical data channel (DPDCH) format and / or an extended dedicated physical data channel (E-DPDCH) format.

The concept disclosed herein can be applied to a WCDMA system having two different interlace structures, for example a 2-ms transmission time interval and a 10-ms transmission time interval. Thus, the front end memory, demodulator and subtractor may be configured to span one or more subpackets of packets having different transmission time intervals.

For TIC, traffic data may be transmitted by one or more users in at least one of the EV-DO Distribution 0 format or the EV-DO Rev A format.

The specific decoding order described herein may correspond to the demodulation and decoding order. Since the process of demodulating packets from the FERAM 312 shifts interference cancellation to a better decoder input, recoding of the packets must occur from the re-demodulation. Repetitive Interference Cancellation by Iterative Finger Delay Adaptation

The ability to accurately estimate the multipath channel and reconstruct the contribution of interference from one user can have a significant impact on the system performance of a CDMA receiver implementing an interference cancellation (IC) scheme.

If the packet is successfully decoded (e.g., passes the CRC), the data symbols are typically sent at a much higher power level than the pilot symbols, so to provide a much more accurate channel estimate for the predetermined finger delay (pilot). Data symbol based channel estimation (DBCE) may be utilized. However, the DBCE algorithm is a finger-specific estimation approach and does not cover how to find the delay offsets for different multipaths.

In a real multipath environment, there are usually multiple physical paths that reach within one chip. Typically, the DBCE estimates the channel coefficients at the finger offsets specified in the demodulation data path, where the fingers are typically assigned at least one chip apart offset to prevent the fingers from correlating. However, once the data is known, the receiver has the freedom to place the finger in any position for interference cancellation.

In addition, some IC schemes may independently estimate channel coefficients of each multipath of a user, and then reconstruct / delete the interference of each of the multipaths in parallel. However, if two multipaths are separated from each other in a pair of chips, the energy of one multipath side lobe or main lobe may leak to the other multipath main lobe, and the two multipaths carry the same data. This multipath correlation (MPC) phenomenon leads to a consistent bias for the estimated channel values. This effect is shown in FIG. 28 based on the ITU pedestrian B channel model. 28 shows a graph of the actual and reconstructed channel impulse response (CIR). In FIG. 28, circles represent channel estimates that are very consistent with the peaks of the actual CIR. However, the reconstructed CIR is raised (compared to the actual CIR) due to the bias.

The present disclosure describes an iterative IC by an iterative finger delay adaptation algorithm, which can better estimate the overall multipath channel impulse response and thus eliminate interference more efficiently.

)에서, 수신기는 신호대 잡음+간섭비(SINR)(

)를 추정할 수 있다. L'는 할당되는 총 핑거 수를 지시하고, 이는 동일한 사용자의 총 다중 경로 수와 같지 않을 수도 있다. 그 다음, 수신기는 핑거들을

의 세기와 관련하여 가장 강한 것에서부터 가장 약한 것으로 리스트에 정렬한다.29A illustrates an iterative IC method with iterative finger delay adaptation in accordance with an aspect of the present invention. At block 2900, the repetition of the receiver is set to 1 (repeat = 1). At block 2902, each predetermined finger delay (e.g., from a time tracking loop of the demodulation process)

In the receiver, the signal-to-noise + interference ratio (SINR) (

) Can be estimated. L 'indicates the total number of fingers allocated, which may not be equal to the total number of multipaths of the same user. The receiver then picks up the fingers

Sort in the list from the strongest to the weakest with respect to the strength of.

라면, 수신기는 이 핑거에 대한 삭제를 시도하지 않는다.At block 2904, the receiver performs continuous channel estimation (CE) and interference cancellation (IC) on the fingers in the order in the list, that is, the next finger's interference until the interference of the previous finger is removed from the sample buffer. Channel estimation does not begin. Also, for one finger

If not, the receiver does not attempt to delete this finger.

를 결정하고, 만족한다면 수신기는 블록(2908)으로 진행하고, 그렇지 않으면 수신기는 블록(2922)에서 반복적인 IC 프로세스를 종료한다.At block 2906, the receiver is

And the receiver proceeds to block 2908 if satisfied, otherwise the receiver terminates the iterative IC process at block 2922.

At block 2908, the repetition is incremented by 1, that is, the repetition is set to "Repeat + 1". At block 2910, if repetition> MaxItNum, the receiver proceeds to block 2912, otherwise the receiver proceeds to block 2914.

At block 2912, the reconstructed interference is accumulated for all fingers and removed from the received signals. The receiver terminates the repetitive IC at block 2922.

At block 2914, for each finger 1 = L ', the receiver first subtracts the interference from all other fingers based on the channel estimates at the previous iteration or the latest available channel estimates from all received signals. .

) 및

의 각 사이드에 대한 M개의 이웃하는 오프셋(chipx8 해상도)에서, 수신기는 2M+1개의 오프셋 각각에서 채널 추정치의 진폭을 계산한다. 블록(2918)에서, 수신기는 새로운

을 (총 2M+1개의 오프셋 중) 가장 큰 채널 추정 진폭을 갖는 오프셋으로서 설정한다.In block 2916, for the same finger, the finger delay used in the previous iteration (

) And

At M neighboring offsets (chipx8 resolution) for each side of the receiver, the receiver calculates the amplitude of the channel estimate at each of 2M + 1 offsets. At block 2918, the receiver is new

Is set as the offset with the largest channel estimate amplitude (of 2M + 1 offset total).

에서 수신기는 이 핑거에 대한 CE 및 IC 알고리즘을 수행한다.At block 2920, the new

The receiver performs CE and IC algorithms on this finger.

를 얻기 위해 데이터 심벌들을 기초로 채널 추정을 수행하며, 최소 제곱 에러(MSE)를 최소화하기 위해

값에는 스케일링 팩터가 곱해진다. α는 다음과 같이 선택된다:

식 21 여기서 γ는 핑거별 SINR을 지시하고, N은 상기

값을 얻는데 기초가 되는 누적되는 데이터 심벌들의 길이 를 나타낸다.In the above channel estimation (CE) step, for each predetermined finger, the receiver

Perform channel estimation based on the data symbols to obtain the equation, and to minimize the least squares error (MSE)

The value is multiplied by the scaling factor. α is selected as follows:

Equation 21 wherein γ denotes a finger-specific SINR, and N denotes the above

It represents the length of the cumulative data symbols on which the value is obtained.

Procedures from blocks 2914-2920 are repeated for all fingers, with the receiver recognized as returning to block 2908.

At block 2922, the iterative IC process ends.

29B illustrates an apparatus that includes means 2930-2952 to perform the method of FIG. 29A. The means 2930-2952 of FIG. 29B may be implemented in hardware, software, or a combination of hardware and software.

There are two important parts of this algorithm: one is an attempt to bias the estimated channel coefficients from another multipath due to side lobe energy leakage; The other is to improve the estimated finger delay and potentially place more than one finger in one chip. Information about the predetermined finger delay typically comes from the demodulation data path. Therefore, finger delays are usually more than 1 chip apart from each other. In addition, this algorithm can prevent locking on side lobes of one very strong path.

30 shows a graph of actual, reconstructed and improved CIRs. 30 demonstrates that the improved reconstructed CIR matches much better with the actual CIR based on the proposed new algorithm.

The algorithm described above assumes that the same number of fingers are fixed throughout the iteration. With some increased receiver complexity, the receiver may find more fingers after removal of existing fingers. If the newly identified fingers reach certain intensity thresholds, the receiver may add them to the tracked finger list with an iterative approach, thus further improving IC performance.

The approaches described here can be applied to both CDMA RL (multiple access) and FL (nested coding) and are suitable for both CDMA2000 ^® groups and WCDMA groups.

The iterative interference cancellation approach described above may enable correction of finger delay and channel estimates at each finger delay. This approach allows the receiver to place more than one finger in one chip. This approach can separate the finger delay estimation problem for the demodulation and interference cancellation data paths.

The approach includes (1) removing bias for channel estimates due to multipath side lobe leakage; (2) allow the receiver to place more than one finger in one chip; (3) prevent locking to side lobes of one strong path; (4) Providing a more reliable route can provide better handling of fat-path channels (eg, ITU pedestrian B, vehicle A channels).

Those skilled in the art will understand that information and signals may be represented using any of a variety of other techniques and methods. For example, data, commands, instructions, information, signals, bits, symbols, and chips that may be referred to throughout the description may include voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any of these. It can be expressed in combination.

Those skilled in the art recognize that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the forms disclosed herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate this hardware and software compatibility, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented in hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the invention.

The various illustrative logic blocks, modules, and circuits described in connection with the forms disclosed herein include general purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays designed to perform the functions described herein. FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microprocessor, or state machine. A processor may also be implemented in a combination of computing devices, eg, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other configuration.

In addition, the steps of a method or algorithm described in connection with the forms disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or other known storage media. The storage medium is coupled to the processor such that the processor can read information from and write information to the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

Headings are included herein by reference and to assist in locating specific sections. These headings do not limit the scope of the concepts described below, and the concepts may be applied to other sections throughout the entire specification.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other forms without departing from the spirit or scope of the invention. Thus, the present invention is not limited to the forms and / or features shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (29)

An interference cancellation (IC) method in a communication system, Receiving multiple paths of a signal; And Performing iterative interference cancellation to remove multipath interference, Performing iterative interference cancellation to remove the multipath interference, Estimating a signal-to-interference + noise ratio (SINR) at each of the plurality of predetermined Rake receiver finger delays; Ordering fingers in a list in view of the strength of the estimated SINR; And Performing continuous channel estimation (CE) and interference cancellation on the Rake receiver fingers in order in the list, Wherein the channel estimation of the next finger does not begin until the interference of the previous finger is removed from the sample buffer. delete delete delete The method of claim 1, If the iteration is not greater than a predetermined maximum iteration number (MaxItNum), For each finger l , subtracting interference from all other fingers reconstructed in the previous iteration from the total received signals; For the same finger, the finger delay used in the previous iteration (

), And

Calculating the amplitudes of the channel estimates at each of these neighboring offsets, at M neighboring offsets on each side of the; The offset with the largest channel estimate amplitude of all the offsets

Setting to; And Remind new

Further comprising: performing a CE and IC algorithm for the finger. The method of claim 5, Performing the CE and IC algorithms is performed by multiplying the scaling factor α to minimize the mean squared error (MSE).

Performing channel estimation based on data symbols to obtain a value. The method of claim 6, α is

Is selected as γ represents the finger-specific SINR, N is the above

A method of interference cancellation that is the length of cumulative data symbols upon which a value is obtained. The method of claim 5, Performing iterative interference cancellation to remove the multipath interference includes removing bias of estimated channel coefficients from other multipaths due to side lobe or main lobe energy leakage. . delete The method of claim 1, If the repetition is not greater than a predetermined maximum iteration number MaxItNum, removing the reconstructed interference accumulated for all fingers from the received signals. The method of claim 1, And the signal comprises a code division multiple access (CDMA) or wideband CDMA (WCDMA) signal. Interference Cancellation (IC) device, A receiver configured to receive multiple paths of a signal; And A module configured to perform repetitive interference cancellation to remove multipath interference, The module, Means for estimating a signal-to-interference + noise ratio (SINR) at each of a plurality of predetermined Rake receiver finger delays; Means for ordering fingers in a list in terms of the strength of the estimated SINR; And Means for performing continuous channel estimation (CE) and interference cancellation on the Rake receiver fingers in order in the list, Wherein the channel estimation of the next finger does not begin until the interference of the previous finger is removed from the sample buffer. delete delete delete 13. The method of claim 12, If the iteration is not greater than the predetermined maximum number of iterations, Means for each finger l , subtracting interference from all other fingers reconstructed in the previous iteration from the total received signals; For the same finger, the finger delay used in the previous iteration (

), And

Means for calculating amplitudes of channel estimates at each of these neighboring offsets, at M neighboring offsets on each side of the; The offset with the largest channel estimate amplitude of the total offsets is new.

Means for setting to; And Remind new

And means for performing a CE and IC algorithm for the finger. 13. The method of claim 12, The means for performing the CE and IC algorithms is multiplied by a scaling factor α to minimize the mean squared error (MSE).

And further configured to perform channel estimation based on the data symbols to obtain a value. The method of claim 17, α is

Is selected as γ represents the finger-specific SINR, N is the above

12. An interference cancellation apparatus which is the length of accumulated data symbols upon which to obtain a value. 13. The method of claim 12, Wherein the module is further configured to remove bias of estimated channel coefficients from other multipaths due to side lobe or main lobe energy leakage. delete 13. The method of claim 12, And if repetition is not greater than a predetermined maximum number of repetitions MaxItNum, further comprising means for removing reconstructed interference accumulated for all fingers from the received signals. 13. The method of claim 12, And the signal comprises a code division multiple access (CDMA) or wideband CDMA (WCDMA) signal. As a base station, A memory configured to store samples of multiple paths of signals received from multiple access terminals; The stored samples are demodulated using a first code sequence corresponding to a first access terminal, and the scaling factor α is multiplied to minimize the mean squared error (MSE).

A demodulator configured to perform channel estimation based on data symbols to obtain a value; A decoder configured to decode data from the demodulated samples; A reconstruction unit configured to use decoded data to reconstruct the encoded and modulated samples of the multipaths; And A repetitive interference cancellation unit configured to remove the reconstructed samples of the multipaths from samples stored in the memory according to any one of claims 12, 16-19, 21 and 22. Base station. The method of claim 23, The demodulator comprises a rake receiver having a plurality of finger processing units for processing multiple paths, each finger having an inherent delay for processing samples from the memory. The method of claim 23, And the reconstruction unit is configured to reconstruct the data by re-encoding, reinterleaving, remodulation, reapplying and respreading of the data channel gains. delete The method of claim 23, α is

Is selected as γ represents the finger-specific SINR, N is the above

A base station that is the length of cumulative data symbols upon which a value is obtained. The method of claim 23, And the base station is configured to receive and process code division multiple access (CDMA) signals from the access terminals. The method of claim 23, And the base station is configured to receive and process wideband CDMA (WCDMA) signals from the access terminals.

Images